MiniMax H3 BF16: Compatible GPUs, VRAM and Download

Text encoder · BF16 · 51.51GB · updated 2026-09-28

Item	Value
File name	`qwen3vl_32b_minimax_h3_bf16.safetensors`
Component	Text encoder
Quant / format	BF16
Size	51.51 GB
ComfyUI folder	text_encoders (text encoder folder)
Source	Comfy-Org/MiniMax-H3

Which GPUs can run it

736 × 416 · 5s: 46 GPUs workable

GTX 1060 6GB GTX 1070 8GB GTX 1080 8GB GTX 1080 Ti 11GB RTX 2060 6GB RTX 2060 12GB RTX 2060 Super 8GB RTX 2070 Super 8GB RTX 2080 Ti 11GB RTX 2080 Ti 22GB (modded) RTX 3050 8GB RTX 3060 8GB RTX 3060 12GB RTX 3060 Ti 8GB RTX 3070 8GB RTX 3070 16GB (modded) RTX 3070 Ti 8GB RTX 3080 10GB +28 more

1344 × 768 · 5s: 46 GPUs workable

1344 × 768 · 15s: 35 GPUs workable

RTX 2060 12GB RTX 2080 Ti 11GB RTX 2080 Ti 22GB (modded) RTX 3060 12GB RTX 3060 Ti 8GB RTX 3070 8GB RTX 3070 16GB (modded) RTX 3070 Ti 8GB RTX 3080 10GB RTX 3080 12GB RTX 3080 20GB (modded) RTX 3080 Ti 12GB RTX 3090 24GB RTX 4060 Ti 8GB RTX 4060 Ti 16GB RTX 4070 12GB RTX 4070 Super 12GB RTX 4070 Ti 12GB +17 more

“Workable” = estimated status ok or warn with 32GB RAM / fast-disk (warn means weight offloading and noticeably slower).

Every quant in this family, side by side

Same family = different quants of the same base variant. They are interchangeable — what changes is size, speed and quality loss. "Peak" is a rough VRAM estimate (file size × 0.6, floor 8G).

File	Quant	Size	Pruned	VRAM fit
GGUF Q2	GGUF Q2	8.49 GB (peak ~8G)	—	runs on 8GB cards
GGUF Q2	GGUF Q2	13.1 GB (peak ~8G)	—	for 12-16GB cards
NVFP4 AWQ	NVFP4	15.69 GB (peak ~9G)	—	for 12-16GB cards
W4A8	W4A8	15.7 GB (peak ~9G)	—	for 12-16GB cards
GGUF Q4	GGUF Q4	18.22 GB (peak ~11G)	—	wants 24GB+
INT8	INT8 ConvRot	27.14 GB (peak ~16G)	—	wants 24GB+
BF16 ← this page	BF16	51.51 GB (peak ~31G)	—	wants 24GB+

Rule of thumb: anything under 10GB fits an 8GB card, the 10-16GB tier targets 12-16GB cards, and files above 16GB want 24GB+. Within this family the flagged NVFP4 AWQ is the official default — take it unless you have a reason not to. The smallest here is GGUF Q2 (8.49GB) and the largest is BF16 (51.51GB); this file (51.51GB) sits in the "wants 24GB+" tier. About this quant: the full-precision baseline: highest quality ceiling, biggest files — mostly for large-VRAM, max-quality setups.

Who is this file for?

Role in the pipeline: the text encoder — it turns prompts into condition vectors and loads on every run; downgrading the TE quant usually hurts less than downgrading the main model.Frankly, a 51.51GB file is all but guaranteed to run through weight offloading unless you have 40GB+ of VRAM. Most people should grab a 10-25GB pruned INT8 or W4A8 build from the table above instead (this family's W4A8 at 15.7GB are a fraction of the size); this one is for "VRAM to spare, full precision only" setups. Full-precision weights also read from disk more slowly, so expect a longer first load.

Fastest vs. minimum hardware

"Fastest" = the card with the shortest estimated time; "floor" = the smallest-VRAM card that still runs it. Times include model load and text encoding; the first run also compiles kernels.

736 × 416 · 5s: fastest RTX 5090 32GB at ~1 min 2 s (46/49 GPUs workable) · floor GTX 1060 6GB (running right at the VRAM ceiling)
1344 × 768 · 5s: fastest RTX 5090 32GB at ~2 min 19 s (46/49 GPUs workable) · floor GTX 1060 6GB (running right at the VRAM ceiling)
1344 × 768 · 15s: fastest RTX 5080 16GB at ~11 min 38 s (35/49 GPUs workable) · floor RTX 4060 Ti 8GB (running right at the VRAM ceiling)

Download, place, refresh — 3 steps

Grab it: the button above redirects to the official HuggingFace / GitHub repo — /go is just a redirect, we never rehost. Skip cloud-drive mirrors; if HuggingFace is slow or blocked on your network, a mirror can speed up the transfer, but verify the file against the official repo. Keep at least 2× the file size free on disk.
Place it: drop the file into text_encoders (text encoder folder) exactly as shown in the spec table, filename untouched — loaders locate files by folder and name, so a rename makes it "vanish".
Load it: hit Refresh in ComfyUI (it only scans folders at startup or on refresh), then pick the file in the loader node. loader node: CLIPLoader — pick the type your workflow specifies.

Sanity check: once downloaded, compare the size on disk with the 51.51GB listed here. Way off? The download is incomplete — browsers sometimes keep a half-written file looking like a finished one — so delete it and pull again.

Also required: The text encoder renders nothing on its own: pair it with a main model (P- INT8; pick the ref2va build if that is your base) and the VAE files.

Won't load? Check these three

Nothing shows up after Refresh: nine times out of ten it's the wrong folder — re-check the ComfyUI folder row in the spec table above. The file must sit inside ComfyUI's models/ subfolder, and no, a desktop shortcut doesn't count.
Instant OOM the moment it loads: make sure no other big model is sitting in VRAM first, then step down to a smaller quant from the table above — the INT8 vs BF16 guide has the full math on why brute-forcing backfires.
Renamed it and now nothing picks it up: loaders match by filename (GGUF metadata stores the original name too), so restore the original name and hit Refresh once more.

Download · HuggingFace →

→ Run the numbers for your RAM/budget in the calculator (this file pre-filled)

FAQ

How much VRAM does qwen3vl_32b_minimax_h3_bf16.safetensors need?

The file is 51.51GB (Text encoder). Runtime VRAM also depends on resolution and duration; a 720p-class short clip typically needs roughly 31GB+ of free VRAM. Run the calculator for your exact card.

Where does this file go in ComfyUI?

text_encoders (text encoder folder). Keep the filename as-is; refresh ComfyUI and select it in the matching loader.

How does it compare with other quantizations?

See sibling entries in the model library and the INT8/BF16 comparison article. Rule of thumb: 8GB → GGUF Q3/Q2; 12-16GB → pruned INT8 or W4A8; 24GB+ → full INT8 or NVFP4 (RTX 50).

Similar files: GGUF Q2 · INT8 · NVFP4 AWQ · W4A8 · GGUF Q2 · GGUF Q4